Why AWS public cloud customers rely on Intel Xeon Scalable processors

Amazon EC2 on IntelAmazon Web Services (AWS) has deployed Intel Xeon Scalable processors to ensure better performance to its public cloud customers when they opt for Amazon EC2 C5 instances.

Intel said the recently launched Xeon Scalable processors are architected for evolving cloud data center infrastructure, offering energy efficiency and system-level performance that average 1.65x higher performance over the prior generation Xeon processors.

AWS has optimized AI / deep learning engines with the latest version of the Intel Math Kernel Library and the Intel Xeon Scalable processors to increase inference performance by over 100x on the new Amazon EC2 C5 instances.

The Intel Xeon processors’ architectural features boost the efficiency and performance of running deep learning training and inference workloads in cloud environments. Intel has optimized MXNet and other deep learning frameworks to run on Amazon EC2 C5 instances.

Intel Xeon Scalable processors deliver 2.4x higher deep learning inference performance and 2.2x higher deep learning training performance compared with prior generation.

HPC workloads running on Amazon EC2 C5 instances will increase the speed of research and reduce time-to-results. Intel Xeon Scalable processors feature Intel Advanced Vector Extensions-512 (Intel AVX-512) instructions that make them an ideal solution for compute-intensive scientific modeling, financial operations, and distributed analytics that require high-performance, floating point calculations.

C5 instances is the next generation of compute optimized instances for Amazon Elastic Compute Cloud (Amazon EC2).

Amazon EC2 C5 instances include up to 72 vCPUs (twice that of previous generation compute-optimized instances), 144 GiB of memory, and a base clock frequency of 3.0 GHz to run the most demanding HPC workloads.

“C5 instances set a new standard for consistent, high-performance cloud computing, eliminating practically any virtualization overhead through custom AWS hardware, and delivering a 25 percent improvement in compute price-performance over C4 instances—with some customers reporting improvements of well over 50 percent,” said Matt Garman, vice president, Amazon EC2, AWS.

Netflix, an internet television network with 104 million members in over 190 countries enjoying more than 125 million hours of TV shows and movies per day, has tested Amazon EC2 C5.

“In our testing, we saw significant performance improvement on Amazon EC2 C5, with up to a 140 percent performance improvement in industry standard CPU benchmarks over C4,” said Amer Ather, cloud performance architect at Netflix. “The 15 percent price reduction in C5 will deliver a compelling price-performance improvement over C4.”

iPromote, a provider of digital advertising solutions to 40,000 small and medium-sized businesses (SMBs), processes billions of ad serving bid transactions every day.

“During testing, C5 instances improved our application’s request execution time by over 50 percent and significantly improved our network performance overall,” said Matt Silva, COO at iPromote.

Grail, a life sciences company, processes DNA sequencing data to detect faint tumor DNA signals in a sea of background noise.

“We are eager to migrate onto the AVX-512 enabled c5.18xlarge instance size. With this change, we expect to decrease the processing time of some of our key workloads by more than 30 percent,” said Cos Nicolaou, head of Technology at Grail.

Alces Flight Compute makes it easy for researchers to spin up High Performance Computing (HPC) clusters of any size on AWS.

“With the support for AVX-512, the new c5.18xlarge instance provides a 200 percent improvement in FLOPS compared to the largest C4 instance,” said Wil Mayers, director of Research and Development for Alces.

“This will reduce the execution time of the scientific models that our customers run on the Alces Flight platform. The larger c5.18xlarge size with 72vCPUs reduces the number of instances in the cluster, and has a direct benefit for our user base on both price and performance dimensions,” Wil Mayers said.

Rescale enables customers in the aerospace, automotive, life sciences and energy sectors to run utility supercomputers using AWS.

“C5 fully supports NVMe and is ideal for the I/O intensive HPC workloads seen on Rescale’s ScaleX platform,” said Ryan Kaneshiro, chief architect at Rescale. “C5’s higher clock speed and AVX-512 instruction set will allow our customers to run their CAE simulations significantly faster than on C4 instances.”